I noticed recently that the SSDs hosting the ZIL for my pool had a large
number in the SMART attribute for total LBAs written (with some
calculation, it seems to be the total amount of data written to the pool so
far), did some testing, and found that the ZIL is being used quite heavily
(matching
-0500, Timothy Coalson wrote:
client: ubuntu 11.10
/etc/fstab entry: server:/mainpool/storage /mnt/myelin nfs
bg,retry=5,soft,proto=tcp,intr,nfsvers=3,noatime,nodiratime,async
0
0
nfsvers=3
NAME PROPERTY VALUE SOURCE
mainpool/storage sync standard
Carosone wrote:
On Wed, Jun 13, 2012 at 05:56:56PM -0500, Timothy Coalson wrote:
client: ubuntu 11.10
/etc/fstab entry: server:/mainpool/storage /mnt/myelin nfs
bg,retry=5,soft,proto=tcp,intr,nfsvers=3,noatime,nodiratime,async 0
0
nfsvers=3
NAME PROPERTY VALUE
The client is using async writes, that include commits. Sync writes do not
need commits.
Are you saying nfs commit operations sent by the client aren't always
reported by that script?
What happens is that the ZFS transaction group commit occurs at more-or-less
regular intervals, likely 5
with the current behavior (and the SSDs shouldn't give out
any time soon even being used like this), if it isn't possible to
change it.
Thanks for all the help,
Tim
On Thu, Jun 14, 2012 at 10:30 PM, Phil Harman phil.har...@gmail.com wrote:
On 14 Jun 2012, at 23:15, Timothy Coalson tsc...@mst.edu wrote
, is it smart
enough to do this?
Tim
On Fri, Jun 15, 2012 at 10:56 AM, Richard Elling
richard.ell...@gmail.com wrote:
[Phil beat me to it]
Yes, the 0s are a result of integer division in DTrace/kernel.
On Jun 14, 2012, at 9:20 PM, Timothy Coalson wrote:
Indeed they are there, shown with 1 second
On Fri, Jun 15, 2012 at 12:56 PM, Timothy Coalson tsc...@mst.edu wrote:
Thanks for the suggestions. I think it would also depend on whether
the nfs server has tried to write asynchronously to the pool in the
meantime, which I am unsure how to test, other than making the txgs
extremely
On Fri, Jun 15, 2012 at 5:35 PM, Jim Klimov jimkli...@cos.ru wrote:
2012-06-16 0:05, John Martin wrote:
Its important to know...
...whether the drive is really 4096p or 512e/4096p.
BTW, is there a surefire way to learn that programmatically
from Solaris or its derivates
prtvtoc device
Sorry, if you meant distinguishing between true 512 and emulated
512/4k, I don't know, it may be vendor-specific as to whether they
expose it through device commands at all.
Tim
On Fri, Jun 15, 2012 at 6:02 PM, Timothy Coalson tsc...@mst.edu wrote:
On Fri, Jun 15, 2012 at 5:35 PM, Jim Klimov
So I can either exchange the disks one by one with autoexpand, use 2-4 TB
disks and be happy. This was my original approach. However I am totally
unclear about the 512b vs 4Kb issue. What sata disk could I use that is big
enough and still uses 512b? I know about the discussion about the
worst case). The worst case for 512 emulated sectors on zfs is
probably small (4KB or so) synchronous writes (which if they mattered
to you, you would probably have a separate log device, in which case
the data disk write penalty may not matter).
Good to know. This really opens up the
What makes you think the Barracuda 7200.14 drives report 4k sectors? I gave
up looking for 4kn drives, as everything I could find was 512e. I would
_love_ to be wrong, as I have 8 4TB Hitachis on backorder that I would
gladly replace with 4kn drives, even if I had to drop to 3TB density.
I
- Will I be able to buy a replacement in 3-3 years that reports the disk in
such a way, that resilvering will work? According to the Advanced Format
threat this seems to be a problem. I was hopimg to get arond this with these
disks and have a more future proof solution
I think that if you are
I think that if you are running an illumos kernel, you can use
/kernel/drv/sd.conf
That refers to creating a new pool and is good to know.
Two things: one, it looks like you should also be able to trick it
into using 512 sectors on a 4k disk, allowing you to do exactly such a
replacement
As I understand it, the used space of a snapshot does not include anything
that is in more than one snapshot. There is a bit of a hack, using the
verbose and dry run options of zfs send, that will tell you how much data
must be transferred to replicate each snapshot incrementally, which should
Is there a way to get the total amount of data referenced by a snapshot
that isn't referenced by a specified snapshot/filesystem? I think this is
what is really desired in order to locate snapshots with offending space
usage. The written and written@ attributes seem to only do the reverse. I
I went ahead and hacked this script together, so let me elaborate. First,
though, a teaser:
$ ./snapspace.sh mainpool/storage
SNAPSHOTOLDREFS UNIQUE UNIQUE%
zfs-auto-snap_monthly-2011-11-14-18h59 34.67G 11.0G 31%
When I wrote a script for this, I used separate snapshots, with a different
naming convention, to use as the endpoints for the incremental send. With
this, it becomes easier: find the newest snapshot with that naming
convention on the sending side, and check that it exists on the receiving
side.
Unless i'm missing something, they didn't solve the matching snapshots
thing yet, from their site:
To Do:
Additional error handling for mismatched snapshots (last destination snap
no longer exists on the source) walk backwards through the remote snaps
until a common snapshot is found and destroy
On Wed, Sep 12, 2012 at 7:16 PM, Ian Collins i...@ianshome.com wrote:
On 09/13/12 10:23 AM, Timothy Coalson wrote:
Unless i'm missing something, they didn't solve the matching snapshots
thing yet, from their site:
To Do:
Additional error handling for mismatched snapshots (last destination
I think you can fool a recent Illumos kernel into thinking a 4k disk is 512
(incurring a performance hit for that disk, and therefore the vdev and
pool, but to save a raidz1, it might be worth it):
http://wiki.illumos.org/display/illumos/ZFS+and+Advanced+Format+disks , see
Overriding the Physical
On Mon, Sep 24, 2012 at 12:20 AM, Timothy Coalson tsc...@mst.eduwrote:
I think you can fool a recent Illumos kernel into thinking a 4k disk
is 512 (incurring a performance hit for that disk, and therefore the vdev
and pool, but to save a raidz1, it might be worth it):
http
I found something similar happening when writing over NFS (at significantly
lower throughput than available on the system directly), specifically that
effectively all data, even asynchronous writes, were being written to the
ZIL, which I eventually traced (with help from Richard Elling and others
believe that a
failed unmirrored log device is only a problem if the pool is ungracefully
closed before ZFS notices that the log device failed (ie, simultaneous
power failure and log device failure), so mirroring them may not be
required.
Tim
On Wed, Oct 3, 2012 at 2:54 PM, Timothy Coalson tsc
Several times, I destroyed the pool and recreated it completely from
backup. zfs send and zfs receive both work fine. But strangely - when I
launch a VM, the IO grinds to a halt, and I'm forced to powercycle
(usually) the host.
A shot in the dark here, but perhaps one of the disks involved
On Sat, Oct 20, 2012 at 7:39 AM, Edward Ned Harvey
(opensolarisisdeadlongliveopensolaris)
opensolarisisdeadlongliveopensola...@nedharvey.com wrote:
From: Timothy Coalson [mailto:tsc...@mst.edu]
Sent: Friday, October 19, 2012 9:43 PM
A shot in the dark here, but perhaps one of the disks
On Thu, Oct 25, 2012 at 7:35 AM, Jim Klimov jimkli...@cos.ru wrote:
If scrubbing works the way we logically expect it to, it
should enforce validation of such combinations for each read
of each copy of a block, in order to ensure that parity sectors
are intact and can be used for data
Disclaimer: I haven't used LUNs with ZFS, so take this with a grain of salt.
On Fri, Oct 26, 2012 at 4:08 PM, Morris Hooten mhoo...@us.ibm.com wrote:
I'm creating a zpool that is 25TB in size.
What are the recommendations in regards to LUN sizes?
The first standard advice I can give is that
On Sat, Oct 27, 2012 at 9:21 AM, Edward Ned Harvey
(opensolarisisdeadlongliveopensolaris)
opensolarisisdeadlongliveopensola...@nedharvey.com wrote:
From: Edward Ned Harvey (opensolarisisdeadlongliveopensolaris)
Performance is much better if you use mirrors instead of raid.
(Sequential
On Sat, Oct 27, 2012 at 12:35 PM, Jim Klimov jimkli...@cos.ru wrote:
2012-10-27 20:54, Toby Thain wrote:
Parity is very simple to calculate and doesn't use a lot of CPU - just
slightly more work than reading all the blocks: read all the stripe
blocks on all the drives involved in a stripe,
On Wed, Oct 31, 2012 at 6:47 PM, Matthew Ahrens mahr...@delphix.com wrote:
On Thu, Oct 25, 2012 at 2:25 AM, Jim Klimov jimkli...@cos.ru wrote:
Hello all,
I was describing how raidzN works recently, and got myself wondering:
does zpool scrub verify all the parity sectors and the mirror
On Wed, Nov 21, 2012 at 4:45 AM, Eugen Leitl eu...@leitl.org wrote:
A couple questions: is there a way to make WD20EFRX (2 TByte, 4k
sectors) and WD200FYPS (4k internally, reported as 512 Bytes?)
work well together on a current OpenIndiana? Which parameters
need I give the zfs pool in regards
On Wed, Nov 21, 2012 at 1:29 PM, Timothy Coalson tsc...@mst.edu wrote:
On Wed, Nov 21, 2012 at 4:45 AM, Eugen Leitl eu...@leitl.org wrote:
A couple questions: is there a way to make WD20EFRX (2 TByte, 4k
sectors) and WD200FYPS (4k internally, reported as 512 Bytes?)
work well together
On Tue, Dec 4, 2012 at 10:52 PM, Jim Klimov jimkli...@cos.ru wrote:
On 2012-12-03 18:23, Jim Klimov wrote:
On 2012-12-02 05:42, Jim Klimov wrote:
4) Where are the redundancy algorithms specified? Is there any simple
tool that would recombine a given algo-N redundancy sector with
On Sun, Dec 9, 2012 at 1:27 PM, Jim Klimov jimkli...@cos.ru wrote:
In two of three cases, some of the sectors (in the range which
mismatches the parity data) are not only clearly invalid, like
being filled with long stretches of zeroes with other sectors
being uniformly-looking binary data
On Mon, Dec 17, 2012 at 9:36 AM, Truhn, Chad
chad.tr...@bowheadsupport.comwrote:
I am not disagreeing with this, but isn't this the opposite test from what
Ned asked? You can send from an old version (6) to a new version (28),
but I don't believe you can send the other way from the new
On Thu, Dec 20, 2012 at 2:32 PM, Jim Klimov jimkli...@cos.ru wrote:
Secondly, there's 8 vdevs each of 11 disks.
6 vdevs show used 8.19 TB, free 1.81 TB, free = 18.1%
2 vdevs show used 6.39 TB, free 3.61 TB, free = 36.1%
How did you look that up? ;)
zpool iostat -v or zpool list -v
Tim
On Thu, Jan 17, 2013 at 5:33 PM, Peter Wood peterwood...@gmail.com wrote:
The 'zpool iostat -v' output is uncomfortably static. The values of
read/write operations and bandwidth are the same for hours and even days.
I'd expect at least some variations between morning and night. The load on
On Fri, Jan 18, 2013 at 4:55 PM, Freddie Cash fjwc...@gmail.com wrote:
On Thu, Jan 17, 2013 at 4:48 PM, Peter Blajev pbla...@taaz.com wrote:
Right on Tim. Thanks. I didn't know that. I'm sure it's documented
somewhere and I should have read it so double thanks for explaining it.
When in
39 matches
Mail list logo
